Prediction of severe neonatal morbidity using maternal, fetal, social, and delivery factors
Journal of Perinatal Medicine · 23.11.2026
DOI: 10.1515/jpm-2026-0396 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Perinatal Medicine, online erschienen am 23.11.2026. Verfasst von Matthew J Blitz und 7 weiteren Autor:innen. Beteiligte Einrichtung: Northwell, New Hyde Park, NY, USA. Die Arbeit nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objectives
To develop and internally validate prediction models for severe neonatal morbidity (SNM) at admission for delivery and at delivery while evaluating the incremental contribution of maternal, fetal, social, intrapartum, and immediate post-delivery factors.
Methods
We performed a retrospective cohort study of singleton deliveries ≥32 weeks’ gestation at two tertiary hospitals (2019–2023). SNM was defined using a composite of severe neonatal diagnoses and procedures. The cohort was randomly divided into training (70 %) and testing (30 %) cohorts. Multivariable logistic regression models were developed iteratively to predict SNM at admission for delivery and at delivery. Discrimination was assessed using the area under the receiver operating characteristic curve (AUC), compared using the DeLong test, and overall accuracy using Brier scores.
Results
Among 53,872 singleton deliveries, 3,775 newborns (7.0 %) experienced SNM. Maternal factors alone demonstrated moderate discrimination (AUC 0.721, 95 % CI 0.704–0.738); adding fetal and social factors produced a modest increase (AUC 0.738, 95 % CI 0.721–0.755; difference 0.017, 95 % CI 0.005–0.029; p=0.0051). Gestational age was the strongest predictor; others included major structural anomalies, polyhydramnios, abnormal umbilical artery Dopplers, hypertensive disorders, and substance use disorder. Adding intrapartum factors and newborn sex increased the AUC to 0.776 (95 % CI 0.761–0.791); birthweight relative to gestational age added minimal improvement (AUC 0.780, 95 % CI 0.765–0.795).
Conclusions
SNM can be predicted with moderate accuracy using routinely available clinical information. Fetal and social factors modestly improved discrimination, and intrapartum information further improved it, supporting staged neonatal risk assessment during the delivery hospitalization.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Northwell, New Hyde Park, NY, USA.
- Zucker School of Medicine, Hempstead, NY, USA.
- Department of Obstetrics and Gynecology, South Shore University Hospital, Bay Shore, NY, USA.
- Institute of Health Systems Science, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY, USA.
- Department of Pediatrics, Cohen Children's Medical Center, and Department of Occupational Medicine, Epidemiology & Prevention, New Hyde Park, NY, USA.
ORCID-Profile: Matthew J Blitz
Zitieren
APA
Blitz, M. J., Jackson, F. I., Lee, J. Y., Lewis, D., Combs, A., Nimaroff, M., Rochelson, B., & Stroustrup, A. (2026). Prediction of severe neonatal morbidity using maternal, fetal, social, and delivery factors. Journal of Perinatal Medicine. https://doi.org/10.1515/jpm-2026-0396
Vancouver
Blitz MJ, Jackson FI, Lee JY, Lewis D, Combs A, Nimaroff M, et al. Prediction of severe neonatal morbidity using maternal, fetal, social, and delivery factors. Journal of Perinatal Medicine. 2026. doi:10.1515/jpm-2026-0396
BibTeX
@article{Blitz2026,
author = {Blitz, Matthew J and Jackson, Frank I and Lee, Ji Yoon and Lewis, Dawnette and Combs, Adriann and Nimaroff, Michael and Rochelson, Burton and Stroustrup, Annemarie},
title = {Prediction of severe neonatal morbidity using maternal, fetal, social, and delivery factors},
journal = {Journal of Perinatal Medicine},
year = {2026},
doi = {10.1515/jpm-2026-0396}
}
EUVIMED